Gromphas jardim, CUPELLO & VAZ-DE-MELLO, 2015

Cupello, Mario, 2024, The genus Gromphas Dejean, 1836 (Coleoptera, Scarabaeinae): nomenclature, distribution, and conservation, including a contribution to the debate on electronic publications in zoology, Zoosystema 46 (2), pp. 23-59 : 36-44

NEWLY DISCOVERED SPECIMENS OF THE RARE GROMPHAS JARDIM CUPELLO & VAZ-DE-MELLO, 2015

Cupello & Vaz-de-Mello (2015) described G. jardim based on two males and three females from Brazil and Bolivia. These specimens were originally discovered in the collections of the Natural History Museum, London ( NHM), the Oxford University Museum of Natural History ( OUMNH), and the Universidade Federal de Mato Grosso, Brazil ( CEMT). Darren Mann, the OUMNH Coleoptera curator, allowed me to take one of the two OUMNH female paratypes for the museum where I was then based, the Museu Nacional, in Rio de Janeiro (MNRJ) . Unfortunately, three years later, this paratype was destroyed. It was consumed along with c. 5 million other insects housed in the MNRJ by the great fire that devastated the main building of the institution on the evening of 2nd September 2018 ( Escobar 2018). This would have meant that only four specimens are now known to exist in collections, but this is not the case. A few months earlier, in July 2018, I visited the Florida State Collection of Arthropods, in Gainesville, Florida ( FSCA), and there I found a sixth specimen of this rare species, a male ( Fig. 8 ). Then, more recently, after joining Texas A&M University as the new Assistant Curator of Entomology in October 2023, I found in the collection three more specimens, a male and two females. Altogether, rather than decreasing, the number of known specimens has risen to eight, half males, half females.

The TAMU specimens are part of the same series as the OUMNH holotype and paratypes from Bolivia, and originally belonged to the W. D. Edmonds collection; they came to TAMU after Edmonds’ material was donated to the university in 2012 ( Streit 2012). As explained by Cupello & Vaz-de-Mello (2015), all these specimens were misidentified as Gromphas lacordairii (cited as ‘ lacordairei ’) in HamelLeigue et al. ’s (2006, 2009) study of the Bolivian scarabaeine fauna. The male from FSCA, in turn, the second oldest ever recorded for G. jardim , was collected in October 1951 in the town of Trinidad (Cercado, Beni, Bolivia), located only about 50 km from the type locality in San Ignacio de Moxos (Moxos, Beni). It is here addressed in the literature for the first time.

Apart from the body size of the FSCA male – it is the largest specimen so far recorded [total length: 17.2 mm (versus 16.3 mm previous maximum); length without head: 15.0 mm (vs 13); pronotal width: 10.1 mm (vs 9 mm); elytral width: 10.5 mm [not measured before)] –, the newly discovered specimens show no noticeable differences from the previously known material and fit the original description of the species well. The sexual dimorphism that Cupello & Vaz-de-Mello (2015), relying on their small original series, suspected to exist in the size and degree of impression of the posterior pronotal fossae – clearly marked and easily visible to the naked eye in males, much less marked and almost imperceptible in females even under the microscope – is also observed in this new material, suggesting that it is indeed a real sexual difference.

The four new specimens also confirm that most of the differences that Cupello & Vaz-de-Mello listed between G. jardim and G. amazonica , its sister species ( Cupello & Vaz-de-Mello 2015), are indeed accurate, particularly in the shape of the cephalic projection and the apical tubercle of the protibiae, as well as the distinctions concerning colouration and the ventral carina of the protibiae (unarmed in G. jardim , armed with a row of sharp tubercles in G. amazonica ). The only main character then deemed distinct between the two species that I now think does not vary in a non-overlapping way between them is the pronotal granulation. Cupello & Vaz-de-Mello (2015) said that, while in G. jardim the pronotal granulation, in lateral view, reaches the posterior edge of the pronotum, it is ‘absent or rudimentary in [the] posterolateral region after [the] lateral fossa’ in G. amazonica . This description is accurate enough for G. jardim ; all known specimens, including the ones from FSCA and TAMU, have the entire lateral region of the pronotum covered with granules. But a reanalysis of the species showed that G. amazonica is more variable in this regard than previously considered to be. While most of the specimens are indeed smooth after the lateral fossa of the pronotum, a few may show the integument rugose and, for this reason, may be mistaken for G. jardim if no material of the latter is available for comparison. Therefore, I caution readers not to rely solely on this character to tell these species apart. Attention is also needed as to the way that the apical tuft of setae of the male protibiae of G. amazonica is described in Cupello & Vaz-de-Mello’s identification key. The form described refers to large- and medium-sized individuals only; in small males, the tuft tends to be separated from the apical tubercle in a similar way as in G. jardim of all sizes.

ADDITIONAL MATERIAL EXAMINED. — Bolivia. Beni • 1♂; Cercado, Trinidad (‘ BOLIVIA - REGION / AMAZONICA / TRINIDAD’); X.1951; no collector; ex E. N. Kellesvig-Waering collection; FSCA • 1 ♀; Moxos, San Ignacio de Moxos , Río Ichiguita ; 15°08’S, 65°18’W; 155 m; 18. V.2005; C. Hamel and T. Vidaurre leg; human faeces trap, savanna; ex W. D. Edmonds colletion; TAMU • 1♂, 1 ♀; Moxos, San Ignacio de Moxos , Río Ichiguita ; 15°08’S, 65°18’W; 155 m; 19. V.2005; C. Hamel and T. Vidaurre leg; human faeces trap, savanna; ex W. D. Edmonds colletion; TAMU .

ON THE PUBLICATION DATE OF GROMPHAS JARDIM CUPELLO & VAZ-DE-MELLO, 2015

In a previous publication ( Vaz-de-Mello & Cupello 2018a), I followed Dubois et al. ’s (2013; 2015a, b) arguments and considered that an electronic work is only available for nomenclatural purposes once it is published with the pagination and other bibliographical information of its version issued as part of a journal’s volume. For that reason, electronic-only versions of articles that are originally published with their own pagination but are later (re)issued as part of a particular volume of a journal, and, consequently, are given new pagination corresponding to the sequence of works published in that volume, were previously considered unavailable in their ‘detached’ versions. Nomenclatural novelties contained in them ‒ i.e., new names and nomenclatural acts ( Dubois et al. 2013) ‒ would only be given availability when their volume-integrated versions finally appeared. Nevertheless, after additional thought and reading, I changed my stance on this matter.

My new rationale is that when a work is registered in Zoobank and its first electronic-only ‘detached version’ (i.e., not incorporated into a volume) is clearly intended by its journal to be a proper publication (i.e., not just a ‘preliminary version’, ‘preview’, ‘early version’, ‘uncorrected proof ’), this detached version should be deemed the work where new nomenclatural novelties are given availability. In turn, the later, re-paginated version integrated into a journal’s volume should be interpreted in the same way as second printings or second editions of printed works are; i.e., not as a modified version of a previous ‘unfinished’ work, but instead as a new, independent available version of a work that is already available. The detached version with its own pagination and the one with volume pagination should, therefore, be seen as two independent available works; instead of seeing the pagination of the detached version as preliminary and the one of the volume version as definitive, it should be interpreted that the pagination in the former is definitive and the one in the latter is a new version. If this whole argument is accepted, the early electronic-only detached versions comply with Article 8.1.1 and are thus not to be deemed preliminary versions as referred to by Articles 9.9 and 21.8.3. If they are dated, they can be interpreted as preprints as defined in the Code’s glossary. Provided that they comply with the other demands of Article 8, these preprints, i.e., the detached versions, are available publications and can be the place where nomenclatural novelties are made available.

Evidence that the journal has the intention of effectively publishing the work and not merely posting a preliminary version on its website ‒ i.e., that the journal issued the PDF with the clear ‘purpose of providing a public and permanent scientific record’ of the content of the work (Article 8.1.1 of the Code) ‒ will most usually and clearly come from the journal’s use of the unambiguous expression ‘published online on [date]’ or ‘published on [date]’ instead of alternatives like ‘available on-line since/from [date]’ to refer to the date when the detached version of the work was made public. Likewise, the use of the term ‘Version of Record’ (see Krell 2015) confirms the intention of permanently publishing the work, whereas terms like ‘early view’ may indicate a preliminary version and others such as ‘accepted’ certainly denotes a preliminary version.

This whole argument and my new conclusion are particularly relevant in regard to the paper published by me and Vaz-de-Mello in the Journal of Natural History on a new species and the phylogeny of Gromphas . Its electronic-only detached version was uploaded to the journal’s website on 12th October 2015, whilst its volume-integrated version was sent to print on 15th March 2016 (Henrietta Thomson, Taylor & Francis Group, personal communication, 6 February 2020). Considering that the journal uses the expression ‘published online’ to refer to its 2015 detached version ( Fig. 9A ) and that, according to its Editor in Chief Andrew Polaszek (personal communication, 5th February 2020), this is meant to signify that it is the definitive version of the paper (i.e., in his words, the Version of Record), I follow them and treat the 2015 detached version as definitive (i.e., not ‘preliminary’). It is, therefore, the first available work wherein the species-group name G. jardim Cupello & Vaz-de-Mello appeared and where it was made available. The work’s 2016 volume-integrated version is, in turn, a new edition with modified bibliographical information and no special nomenclatural relevance. In conclusion, the publication date of G. jardim is that of the detached version: 12th October 2015.

A similar, though not identical, proposal was put forward by ICZN Commissioner Frank-Thorsten Krell a few years ago ( Krell2015). In a few words, applying the NISO / ALPSP Recommendations, Krell sees the detached and volume-integrated versions not as a ‘first’ and a ‘second’ independently available edition of a publication as I do here, but as the very same work and, indeed, the very same version of the work, for changes in pagination and information about volume and issue (‘bibliographical metadata’) are not considered as changes in the content. Similar to my proposal, availability is gained upon the publication of what I call here the detached version provided that it is considered the Version of Record by the publisher and is not further modified without warning. I decided not to follow Krell’s exact proposal, however, because I judge that mine avoids several uncertainties associated with his. These include determining whether a journal follows the NISO / ALPSP Recommendations (e.g., do amateur or small museum or society journals follow them?) or whether a Version of Record has been later modified to produce a Corrected or an Enhanced Version of Record without explicit indication, in which case the original detached version becomes unavailable under Krell’s guidelines but not under mine. How can one ever know – and be assured – that a Version of Record will not be later modified without warning and, thereby, make a previously available work unavailable?

Krell’s proposal also creates difficulties during the preparation of lists of references and catalogues. How should one cite a work first issued as a detached version? By the pagination and publication date of this detached version or by those of the later volume-integrated version?Take the case of G. jardim , for instance: both versions consist of 27 pages of text (the PDF of the volume version has an additional unpaginated cover page containing basically bibliographical information), but their pagination is not the same. While the 2015 detached version is paginated [1]-27, the 2016 volume version is paginated [943]-969. The information making the name available, in particular, appears on pages 4 (information on the holotype) and 7-8 (description) of the detached version, and on pages 946 and 949-950 of the volume-integrated version. When citing the name G. jardim in a work, its authorship and year of publication must be given as ‘ Cupello & Vaz-de-Mello, 2015 ’, for 2015 was the year when the name was made available under Krell’s criteria (as well as under mine). But should the paper be dated 2015 or 2016 in the references? If 2015, then this will agree with the year of the authorship of the name as cited in the text; but which pagination to cite in the same reference? If the one of the detached version, then the volume-integrated version will essentially be ignored (just like I propose). But if, instead, the pagination of the volume-integrated version is to be cited, then it will be inconsistent with its publication date being cited as 2015, for this is not the pagination of the PDF published in 2015 (this inconsistency is present, for example, on the Zoobank webpage of Cupello & Vaz-de-Mello’s article; see Fig. 10C ). If, to fix this problem, one decides to cite the article in the references based entirely on the information of the volume-integrated version, including its year of publication (2016), then the inconsistency will be with the year of publication mentioned in the text for the authorship of the name (2015). In one way or the other, considering the detached and the volume-integrated versions as the same as proposed by Krell (2015) is, as aptly pointed out by Dubois et al. (2015b), an artificial procedure and creates a lot of problems. Under my proposal, in contrast, the date and pagination to be cited on the list of references are unambiguous: those of the detached version, as it was there where the name was made available.

Dubois et al. (2015a, b) have already made similar comments and stressed the importance of page citation in taxonomic works, a claim that I fully endorse. The fact that pagination is not regulated by the Code, as Krell (2015) argued, does not make it less relevant for taxonomists and cataloguers or its stability less desirable. On the contrary, anyone who has carried out a taxonomic project, particularly the preparation of catalogues and comprehensive systematic revisions ‒ or, to be more general, anyone who has needed to check in the original publication a quotation or specific information cited by a third author ‒ is aware of the importance of knowing, and being able to cite, the exact and unambiguous place in a publication where a piece of information is to be found.

In contrast to Krell’s (2015), my procedure, I believe, is more straightforward and in line with the way the Code and zoologists have been approaching printed publications for decades: once a work has been published in a way to satisfy the Code’s criteria of availability, that version is the definitive one where new nomenclatural acts are established, no matter how many new modified versions are later reissued or whether the original version is no longer obtainable from the publisher. Concerning the latter point – viz., the detached version not being publicly obtainable after being replaced by the volume-paginated version on the journal’s website –, it could be argued (see, e.g., Dubois et al. 2015b) as a difficulty for implementing my proposal. However, if Krell is correct (and I believe he is) that the later volume-integrated versions are almost always identical to the early detached ones except for their bibliographical content (but see Dubois et al. 2013), then there should be no problem: in practice, these volume-integrated versions can be used as surrogates for the detached versions when these are not obtainable. Nevertheless, if Dubois is correct and many journals do produce volume-integrated versions with additional modifications, then my procedure will safeguard the availability of the early works, which would become unavailable under Krell’s, and avoid the chaos of never knowing whether a work has been – or will be – made unavailable due to modifications included without warning in subsequent versions produced by the publisher.

Despite my new stance on electronic publications, I still consider Dubois et al. ’s (2013, 2015a, b) arguments to be in most parts quite sensible, as indeed recognised by Krell (2015) himself, and should be taken into consideration by readers. I only prefer my new stance over Dubois et al. ’s because I consider mine protects more widely the availability and priority of new names and nomenclatural acts and leaves authors less exposed to bad editorial procedures over which they are usually powerless. Under my proposal, editors have to comply with fewer requirements for a work to be available (e.g., they can produce as many modified versions as they like without threatening the availability of the original work), and so it is easier (and faster) for new names and nomenclatural acts to be available than under Dubois et al. ’s, and all while still respecting the principle of permanency of the zoological literature.

It must be noted that Dubois et al. (2013, 2015a, b) have analysed the alternative that I present here ‒ viz., to treat the detached and volume-integrated PDFs as different works and to deem them independently available ‒, but came to a different conclusion. While they agree with the first part of the argument ‒ that the two versions are distinct publications, not the same as argued by Krell (2015) ‒, they consider that the first version, the detached version, is unavailable because, in their opinion, it fails to comply with Articles 8.1.1 and 9.9. Their point is that because the PDF of the detached version is removed from the journal’s website and replaced with the volume version once the latter is published, this would indicate that the journal regarded the detached version as only preliminary and so that it did not have the ‘purpose of providing a public and permanent scientific record’ as demanded by Article 8.1.1. I disagree.

The term ‘published’, if used by the journal to refer to the detached version, clearly denotes the stated purpose of the editors and publishers that the content of this version is public, citable, and a permanent part of the scientific literature. However, if, in practice, they replace the PDF of the detached version with that of the volume-integrated version and the former thus ceases to be accessible from the journal’s website, this does not deny the original intention (the purpose) behind their action. Rather, this only shows either that the editors are bona fide ignorant of the intricacies of the Code (they believe that replacing the original PDF with another nearly identical, differing only in the bibliographical data, would not interfere with the availability of the original work; i.e., it would not interfere with its original purpose of being a permanent scientific record), or that they simply disagree with Dubois et al. ’s interpretation of what a published work is and that, in their interpretation, replacing the PDFs would not go against Article 8.1.1 (e.g., if they follow Krell’s proposal). The point is not whether the PDF of the detached version remains obtainable from the journal’s website or not, but rather, what the original intention of the editors was in preparing the PDF of the detached version. Article 8.1.1 does not demand a work to be permanently public, but only that it being permanently public has to be the publishers’ original intention1.

Being, in practice, permanently obtainable from the publisher has indeed never been a requirement for a work to be available or to be regarded as a permanent scientific record in zoology. For instance, is the 10th edition of Systema Naturae still obtainable from Lars Salvius? Of course not. It is currently only accessible through original copies in a handful of public and private libraries and a very few antique bookshops around the world, through photocopies or facsimile reproductions, or through digitised copies obtainable online from specialised websites (e.g., Biodiversity Heritage Library, Google Books). This situation is analogous to that of detached versions that have been removed from their respective journals’ websites: they remain in existence only on the computers (digital library) of those who downloaded them before removal, or of those who had the PDF shared with them by colleagues, as well as in physical libraries in the form of house-made printed copies, or stored on academic social networks like ResearchGate or Academia. In all these cases, modern or old, new copies ceased to be obtainable from the original publisher.

A similar thing could also be said of many printed journals published (currently or in the past) by small amateur societies, small museums, or even privately by amateurs around the world, particularly in Europe. Anyone who has worked on the taxonomy of Coleoptera , for instance, knows how difficult it can be to obtain copies of papers published by many of these small journals. While this situation is far from desirable, rarity and difficulty of obtaining have never prevented a printed work from being available. Why should it be the case for electronic publications? Although it would undoubtedly be preferable for journals to maintain the detached version on their websites alongside the volume-integrated version, I see no reason why them failing to do so would be any different from printed works that ceased to be obtainable from the original publisher, nor why considering otherwise would be a natural (or even desirable) interpretation of the Code and its 2012 Amendment ( ICZN 2012). This would only create more uncertainties. So, I disagree with Dubois et al. and argue for the proposal that I put forward above.

Yet, I concede that, sometimes, it can be difficult to ascertain the purpose of a detached version uploaded to a journal’s website. Such ambiguity may trigger endless quarrels about the publication date of a work under the Code, and detract zoologists from the actual object of their studies, animals and their biology. It would greatly serve stability and universality in zoological nomenclature and the science of zoology overall if journals and their publishers stated fully unambiguously their intention as related to Art. 8.1.1 when posting the PDFs of works containing zoological names and new nomenclatural acts on their websites2. Currently, there is nothing in the Code enforcing this. All I can do here is urge editors and publishers to expressly state on the first page of the earliest version of a work that they deem published that it has the ‘purpose of providing a public and permanent scientific record’. This will prevent any doubts about the work’s compliance with Art. 8.1.1 and, provided that the other criteria established in Chapter 3 of the Code are met, ensure its availability. At the same time, to solve this problem definitively in a formal way, I put forward the following proposal to the Commission: in the forthcoming 5th edition of the Code, Article 8.5.2 should be amended as follows:

8.5. Works issued and distributed electronically

To be considered published, a work issued and distributed electronically must

[...]

8.5.2. state the following sentence on its first or last page: ‘This work is issued on [exact date of publication, with day, month, year] for the purpose of providing a public and permanent scientific record as established in Article 8.1.1 of the International Code of Zoological Nomenclature and is registered in ZooBank under number [exact Zoobank registration number]’. Text between square brackets is to be replaced with information relating to the work in question.

Figures

FIG. 8. — The male specimen of the rare Gromphas jardim Cupello & Vaz-de-Mello, 2015 newly discovered in the Florida State Collection of Arthropods, Gaines- ville, Florida, USA. With the destruction of one of the paratypes in the 2018 Museu Nacional fire, it is one of the eight specimens known to survive in collections. Photographs courtesy of Paul Skelley (FSCA).

FIG. 9. — The publication date of Gromphas jardim Cupello & Vaz-de-Mello, part I; A, the webpage of Cupello & Vaz-de-Mello’s article in the Journal of Natural History. Notice that the information mixes data from the detached and the volume-integrated versions. The green arrow points to the publication date of the detached version as indicated by the journal, 12th October 2015, whereas the red arrow points to the bibliographical information concerning solely the volumeintegrated version published in 2016. Note also that the journal uses the expression ‘published online’ (not ‘ ahead of print’, ‘early view’, or the like) to refer to the detached version, denoting that it was not considered ‘preliminary’, but a proper publication; B, the upper part of the first page of the PDF of the detached version. The yellow arrow points to the publication year of the PDF, 2015; the white arrow indicates the ‘article history’. Note that, while the full date, including the day and month, of the submission and acceptance are given (white arrow), only the year of the publication is mentioned (yellow arrow); C, the same page in the PDF of the volume-integrated version. The pink arrow points to the bibliographical data of this version (publication year, volume,number, and pagination), whereas the black arrow shows the updated ‘article history’. Observe that here, the full publication date of the detached version (‘Online’) is given. The full date of the publication of the volume-integrated edition itself, however, is not given anywhere either in the PDF or on the journal’s website. I was able to obtain the date when this version was sent to print, 15th March 2016, through direct inquiry with the publisher.

FIG. 10. — The publication date of Gromphas jardim Cupello & Vaz-de-Mello, part II; A, footer of the first page of the PDF of the detached version showing the Zoobank link of the article (yellow arrow); B, footer of the first page of the PDF of the volume-integrated version. Notice that, although this version was published in 2016 (and is dated as such at the upper part of the same page; see Fig. 6C), the footer is dated 2015 (pink arrow). The gray arrow shows that it is the full date of the detached version that is given in the volume-integrated version; C, the webpage of Cupello & Vaz-de-Mello’s article in Zoobank. Note that even though the publication date stated on the website correctly refers to the detached version (green arrow), the information in the ‘Journal Article’ section is inaccurate (red arrow). It mixes the publication year of the detached edition (‘2015’) with the bibliographical information of the 2016 volume-integrated version (‘50 (15-16): 943-969’). The correct pagination of the 2015 detached version – the one wherein G. jardim was made available – is 1-27.